The present invention relates to a technique of aligning a mask pattern and a photosensitive substrate relative to each other, which is applicable to an exposure system used in a photolithography process in which a mask pattern is transferred onto a photosensitive substrate to produce semiconductor devices, for example. More particularly, the present invention relates to a technique of detecting a mask pattern provided on a photosensitive substrate.
In photolithography processes for producing, for example, semiconductor devices, liquid crystal display devices, thin-film magnetic heads, imaging devices (CCD), magneto-optical discs, etc., an exposure system is used in which an image of a transfer pattern formed on a photomask or reticle (hereinafter referred to collectively as "reticle") is transferred onto a photosensitive substrate, e.g., a wafer or glass plate coated with a photoresist, by a projection exposure method through a projection optical system, or by a proximity exposure method.
In such an exposure system, the reticle and the wafer must be aligned with respect to each other with high accuracy prior to the exposure process. In order to effect the alignment, the wafer has position detecting marks (alignment marks) formed (transferred by exposure) thereon in a previous processing step. By detecting the positions of the alignment marks, the position of the wafer (i.e., the circuit pattern on the wafer) can be accurately detected.
Examples of alignment mark detecting methods include a laser beam scan type detecting method, a laser interference type detecting method, etc., in which the position of an alignment mark is detected by detecting laser light scattered or diffracted by the mark. However, laser light has strong monochromatic properties; therefore, the use of laser light may cause position detection accuracy to be degraded on account of adverse effects such as multiple interference between the photoresist surface and the mark surface.
There is another position detecting method in which an alignment mark is illuminated by a broad-band light beam from a light source, e.g., a lamp, and the illuminated mark is imaged through an image-forming optical system to detect the position of the alignment mark on the basis of an image signal output from the image-forming optical system (this type of detection method will be hereinafter referred to as "imaging type position detection"). In contrast to the above-described detection methods that use laser light, the imaging type position detection has a merit that the position detection is unlikely to be influenced by adverse effects such as the presence of the photoresist.
Under the present circumstances where the patterns of semiconductor integrated circuits and other similar devices become increasingly finer, a process of planarizing (i.e., leveling) the wafer surface has recently been introduced as a process carried out after a film deposition process and before a photolithography process. The planarization process has an effect of improving device characteristics by uniformizing the thickness of a deposited film from which a circuit pattern is to be formed, and also has an effect of improving the adverse effects of the wafer surface unevenness on the line width error of a transferred pattern during the photolithography process.
However, in a detection system wherein position detection is effected on the basis of a height difference between the recessed and projecting portions of an alignment mark or a reflectivity variation at the alignment mark portion on the wafer surface, the planarization process may make it impossible to detect an alignment mark because the height difference between the recessed and projecting portions of the alignment mark is considerably reduced by the planarization. Particularly, in a process for an opaque deposited film (i.e., a metal or semiconductor film), the alignment marks are covered with an opaque film of uniform reflectivity. Therefore, position detection depends only on the height difference between the recessed and projecting portions of the marks. Thus, planarization carried out for an opaque deposited film raises the most serious problem.
Incidentally, "dark-field microscopes" and "phase-contrast microscopes" are known as optical systems for detecting only a "step" portion on a substantially flat object to be detected. One type of dark-field microscope is arranged such that an object to be detected is illuminated with a numerical aperture large enough to prevent direct light from entering the image-forming optical system, and an image is formed through the image-forming optical system by using only light scattered by the object. In another type of dark-field microscope, a circular light-blocking member is provided at a part of an optical Fourier transform plane (pupil plane) in the image-forming optical system with respect to the object to be detected such that the center of the circular light-blocking member coincides with the optical axis of the image-forming optical system, and an aperture stop (.sigma. stop) is provided by which the distribution of illuminating light over a plane in the illumination optical system which is conjugate to the light-blocking member, that is, an optical Fourier transform plane in the illumination optical system with respect to the object to be detected, is restricted within a circle which is in image-forming relation to the circular light-blocking member. As the aperture stop (.sigma. stop), a stop having an annular aperture may be employed.
A typical dark-field microscope is adapted to apply illuminating light to an object to be detected (e.g., a position detection mark on a wafer) and to form an image of only higher-order diffracted light (and scattered light), with zeroth-order diffracted light (regularly reflected light) being blocked, among light reflected and diffracted by the illuminated object. Among the diffracted light, zeroth-order diffracted light contains substantially no information concerning unevenness on the object or reflectivity variation of the object. However, higher-order (first- and higher-order) diffracted light contains such information. Therefore, the dark-field microscope enables steps to be made visible more clearly (with higher contrast) than the ordinary (bright-field) microscopes because the image is formed only from higher-order diffracted light with zeroth-order diffracted light being blocked.
Meanwhile, a typical phase-contrast microscope has a phase-contrast filter provided at a pupil plane of an image-forming optical system. The phase-contrast filter gives a phase difference between zeroth-order diffracted light and diffracted (and scattered) light of other orders. The quantity of higher-order (first- and higher-order) diffracted light generated from a mark pattern of low step is extremely small. However, in the phase-contrast microscope, zeroth-order diffracted light, which is generated in a large quantity, can also be contributed to image formation. Therefore, it is possible to obtain an image which is brighter (higher in light intensity) than in the case of the dark-field microscope. It should be noted that, when the intensity ratio of zeroth-order diffracted light to diffracted light of other orders is extremely high, the image contrast reduces, and therefore, zeroth-order diffracted light may be reduced under certain circumstances.
However, if the conventional dark-field microscope is used to detect a position detection mark on a wafer, not only zeroth-order diffracted light, which is not needed for image formation, but also diffracted light of relatively low order (i.e., useful diffracted light, which contributes to image formation) is blocked, causing the image contrast and fidelity to be degraded.
In a case where a conventional phase-contrast microscope is used to detect a position detection mark on a wafer, not only zeroth-order diffracted light, which is not needed for image formation, but also diffracted light of other orders (i.e., useful diffracted light, which contributes to image formation) is subjected to phase-difference imparting action and light-reducing action, causing the image contrast and fidelity to be degraded.
In view of the above-described problems, an object of the present invention is to provide a position detecting apparatus which is capable of accurately and reliably detecting the position of a position detection mark, even a mark having an extremely small height difference (step) between recessed and projecting portions which constitute the mark.